Harness Hub
All people

Dr John Seach

Posts about agent harnesses on X

Posts on X about Open harnesses, Harness engineering, Evals & benchmarks.

On X

Shown as plain text; “Load on X” loads X's script for that post, which may set cookies.

  • Kimi K3 vs GLM-5.3 for coding ⚔️

    They’re close.
    Neither is the clear winner.

    The independent Artificial Analysis index has them tied at 60.

    The real difference is what you actually ask them to do.

    🔶GLM-5.3 is https://t.co/IF23WrF4HT’s August upgrade of the same ~750B base…

    Open on X

More on the X feed

Sources

  1. x.com/johnseach
  2. Sample post